Senator Leahy put out a statement late yesterday afternoon and we’re still not exactly sure what it means–it could be that DNS-blocking provisions will drop out of the Protect IP Act (PIPA, the older brother of the House’s infamous SOPA), or it could be that they will still be in the bill with a delayed implementation, subject to some sort of “study.” But even if the DNS-blocking provisions are gone for good, it’s worth remembering that this bill is still bad for the Internet, bad for users, and bad for the economy. The bill is still bad in the broader sense that it will impose endless obligations on Internet companies and users without disrupting pirates. Pirates, after all, have already developed technology to counteract the worst that the Congress can do. But even a DNS-free version of the bill (which there’s a good chance we don’t even have) contains specific provisions that would interfere with commerce and the exchange of ideas online. Here are some of the biggest problems that remain in PIPA:
Broad language that could sweep in countless websites. Supporters of PIPA often try to argue that the bill only applies to the “worst of the worst” sites on the Internet. But PIPA’s definition of an “Internet site dedicated to infringing activities” includes sites that merely “facilitate” copyright infringement. This is a new concept in copyright law that could apply to any number of sites, from discussion boards, to social media sites, to software repositories like GitHub. PIPA also applies to “information location tools,” a comically over-broad legal concept that includes things such as a “directory, index, reference, pointer, or hypertext link.” That’s right–PIPA requires “hypertext links” to “remove or disable access to the Internet site associated with the domain name set forth [in a court order]; or not serve a hypertext link to such Internet site.”
A private right of action that would encourage frivolous and troll lawsuits. PIPA includes a new “private right of action”–an expedited process that allows plaintiffs to starve websites of advertising or cut them off from their bank accounts. Plaintiffs only need to demonstrate that a site is “dedicated to infringing activities,” a definition that, as noted above, is very broad and includes more than just the “worst of the worst.” Because of the lower standards plaintiffs that might not be able to win an actual copyright infringement suit can knock sites they don’t like off the Internet.
Incentivizing vigilante action. Today if an MPAA studio or other content conglomerate tells a domain name registrar, a bank, an ad network, or other service provider to stop doing business with some disfavored website, it’s going to be told to go take a hike. Of course companies generally don’t like to get in the business of attacking their customers–it’s bad for business. But also, if they do cut off one of their customers, they might find themselves brought to court by their now-former customer. This gives content bullies a lot less leverage than they’d like. PIPA does away with this by creating a one-way ratchet. If a service provider always does what a studio asks it, it’s immune from lawsuits. But if it ignores a letter from a studio it might find itself accused of being “secondarily liable” for copyright infringement, under stretched interpretation of existing law or even under some newfangled theory of “facilitation.” PIPA thus creates a huge incentive for service providers to cut off their customers first and ask questions never.
The fact that one of the most-hyped parts of PIPA might drop out, after sustained outcry from Internet engineers and users, cybersecurity experts, and political commenters representing every viewpoint, should be enough to show how ill-conceived the entire bill is. Its attempts to micro-manage the Internet and make dozens of industries change their shape to suit the needs of just one are not limited to the censorship of the domain name system. Of course it’s a good thing that Senator Leahy is starting to recognize that his bill has serious flaws. And he’s right that the “positive and negative effects” of controversial provisions should “be studied before implemented.” But he’s wrong if he thinks that the only controversial part of the bill are its DNS components–the entire effort needs to be reconsidered.